Understanding Deep Learning: Application in Rare Event Prediction

This repository is for the book, Understanding Deep Learning: Application in Rare Event Prediction.

The link to the book is here.

Created and maintained by,

Chitta Ranjan, Ph.D.

mailto: [email protected]

LinkedIn: www.linkedin.com/in/chitta-ranjan-b0851911/

Medium: https://medium.com/@cran2367

Chapters

1. Introduction

2. Rare Event Prediction

3. Setup

   a. Getting Started

   b. Setup data

4. Multi-layer Perceptron (MLP)

5. Long and Short Term Memory (LSTM)

6. Convolutional Neural Network (CNN)

7. Autoencoders

   a. Dense Autoencoders

   b. LSTM Autoencoders

   c. Convolutional Autoencoders

   d. Classifier pretrained with convolutional autoencoder

8. Appendix

Setup Data

1. Download the data from: data link
2. Place the data file in the /data/ directory.

Video Lectures

The video lectures of the chapter are published on YouTube.

Chapter 1-2 - Recap

Chapter 3 - Part 1 - TensorFlow Installation - Background

Chapter 3 - Part 2 - Getting Started - TensorFlow on Google Colab

Chapter 3 - Part 3 - Data set for Rare Event Prediction

Chapter 4 - Part 1 - Background

Chapter 4 - Part 2 - Understanding Multilayer Perceptrons

Chapter 4 - Part 3 - Deep Learning and Neural Network

Chapter 4 - Part 4 - Intuition behind Neural Networks

Chapter 4 - Part 5 - MLP Math Operations I

Chapter 4 - Part 5 - MLP Math Operations II

Chapter 4 - Part 6 - Importance of Nonlinear Activation

Chapter 4 - Part 7 - Back Propagation

Chapter 4 - Part 8 - Intermission-Outline of rest of the chapter

Chapter 4 - Part 9 - Curve Shifting for Early Prediction

Chapter 4 - Part 10 - MLP training iteration levels

Chapter 4 - Part 11 - Custom metric in TensorFlow I

Chapter 4 - Part 11 - Custom metric in TensorFlow II

Chapter 4 - Part 12 - Dropout I - Co-adaptation Phenomenon

Chapter 4 - Part 12 - Dropout II - Underlying Concept

Chapter 4 - Part 12 - Dropout III (as) A Regularization Technique

Chapter 4 - Part 13 - Activations I - Importance of Gradients

Chapter 4 - Part 13 - Activations II -Gradient vis-à-vis Learning

Chapter 4 - Part 14 - Vanishing & Exploding Gradient

Chapter 4 - Part 15 - Story of Activation Functions

Chapter 4 - Part 16 - Rules-of-thumb MLP Dense Layer Models

Chapter 4 - Codes I - MLP Modeling - Get ready with Data Scaling

Chapter 4 - Codes II - MLP - First (Baseline) Dense Layer Model

The videos of the next chapters will be published from now until the end of 2021.

---